Solar Install; the on-going saga

I looked around and found all those “camper solar kits” were bundled with low end controllers;
I found a not bad one with a steca controller that has a built in ammeter until I did some more research and learned these are not accurate, nor are the solar charge controller remotes.
If you want to read amps you need a shunt or a clamp meter and a calculator.
I read of one customer complaining to Steca that the ah readouts were misleading and the reply received was that this feature was inaccurate, simply that it would be far more expensive for that feature to work correctly.?!


I settled on a 150watt Chinese panel with a decent warranty that came in at the right price and suited my roof dimensions. I had a look at european panels and speculated that they won't honour their warranties soon as China puts them out of business. Bosch are pullling out of the market, Q cells are being outsourced etc.


First agenda I had is to disguise it. My camper lives in the wild and urban, so I need a way to hide it from prying fingers and gobsh1teology.
Simpliest solution I had was make an unassuming roof box and paint it black outside (stealth) and yellow inside (reflective....white would be more reflective but I like yellow so meh!)


2 Mock up side .JPG


I mounted the box onto an existing roof rack with clamped U-brackets.
I secured the module to the box with adapted domestic solar PV brackets, this required a little angle grinding of the brackets and some padding of the panel. I'm not too fond of the extra ballast on the back of the module but it may be useful later for the tilting mounts.
I put a layer of road noise reducing rubber between the backboard and the module more to meet the bracket requirements than the sound ones.
This was cut from an old inner tube...if you want straight pieces cut from the center not the sides...



Road noise reducing filler.JPG
Back board.JPG
Scaffolding solution.JPG
Rack Box.JPG

I decided to spec. the new system to the alternator because that's the hardest component to change, which means getting flooded deep cycle batteries that can take 14.7volts. However I have a sealed battery now I want to make use of it for the time being.


Wiring mess.JPG


Given this I wanted a versatile solar controller, then I decided I wanted a programmable controller because well that's pretty versatile. I went for the Morningstar Tristar 45 because of this and also because it is the only solar controller that can run wind/hydro and hybrid systems too which I plan on using down the line for other projects.

Tristar patch.JPG


The leads from the controller to the battery are 8mm cores from stranded 125amp single phase TRS cable I got through work. I re-used the rubber shielding after wiring everything as cable trunking wrapped in electricians tape...does the job.


Tidied Cables...ish.JPG


The controller has separate battery sense leads (to avoid reading voltage from a load carrying wire), and remote temperature sensor connected to the battery negative terminal for temperature compensation.

Installed.JPG

As far as I can see PWM charging plays havoc with cheap voltmeters. For accurate readings I think you need an RMS meter. I've observed a 0.5v difference between the two after a minor fluster.

Retail blade fuse holder.JPG: €3 - €5 aprox.
Home made blade fuse holder.JPG: €0.30 aprox.

Goes in series with all negative returns to the leisure batteries and the chassis ground is to be connected to the shunt and not the -ive battery terminal . There's smaller kelvin connections on the sides to take voltage sensing wires, where the main current carrying wires are sent to the main lugs on the top. I wired the -ive solar controller battery sense lead onto the kelvin connection, I'm not entirely sure if this is right but it's made little difference.

Shunt Nearside.JPG
Shunt Farside.JPG


The saga isn't going to be as ongoing as I first predicted because, I had intended to do another few posts on refitting my sealed battery for 2 x 6volt flooded golf cart batteries in series, that weren't going to fit where I wanted them, but a few things have transpired since to make me re-evaluate.
I was going to up-rate my mains charger too for one that conforms to the battery manufacturer's spec.



A: My lone battered and bruised battery with solar is outperforming what two new of it's kind were doing with an alternator subsidised weekly mains charge. Oh I don't have to use the alternator anymore to top up the leisure batteries...but I still use the link switch to top up the starter off solar now.
B: I want to see if Morningstar's claims that PWM charging can recover lost battery capacity is true.
C: I don't want to throw out a working battery albeit it working very badly as a simple ecological issue.
I bought a car with my new battery budget and still have €40 change. :cool:
E: Winter will be the true teller if I need to invest given the less light more demand equation but I've already done three Winters in a camper so I'm kinda determined to go bricks and mortar next one.


I'll post some definitive figures where I get them but it might be a while.

tehehe yeah I just put one on a car today to save another starter from the recycle heap. They're fairly rubbish. All it'll do is prevent a battery from self-discharge in the height of summer. We had it tilted 40° directed at the mid-day sun today at 20° Celcius and it took a battery from 12.5volts to 12.9volts producing < 100mA.

For that sort of money you should be considering something like this http://www.hollybrookps.co.uk/pv_modules/80w-solar-panel-monocrystalline.html

That Sealy link is just a trickle charger, it might give you enough for a mobile phone charge. Aside from this the wires are too small and it's unregulated. Not that 100mA could harm a leisure battery at all but the principle of unbridled charge upsets me.

Ah..no, my link is low end, it's a Chinese job...but value for money.

I try to avoid Maplins, they're rip-off merchants. Even though I post lots of links to them. It's a convenience thing. Definitely for solar they're triple what you ought to be paying.

The competitive price per watt on solar modules today is about €1.20 You can go lower or higher for brands, guarantees and warranties.
Also a typical warranty on a solar module is 20 - 25 years...think economy of scale.

The link I posted is portable, there's also smaller ones on that web-site. They have connecting leads on the back you can mount them outside your camper on brackets or rocks or whatever and run a temporary lead to a cheap charge controller near your batteries. You can store them in the back of a wardrobe when you don't want them. They're quite durable.

I really wouldn't consider anything less than 50watts for light camper use. That briefcase will leave you high and dry on a cloudy day, actually at 13watts probably on a sunny day too.

Sorry I haven't posted any figures from my system to compare yet. I've not been using my camper much and won't be until next month. Mostly working on a car these days. I'll post what I can this week but daily averages could be mid July.

The Trimetric Rocks ,

Nominal system voltage 12.6volts (lower the voltage = higher the amperage), with solar module and controller off (night-mode)

• System idle self-consumption (ghost loads & Trimetric); 300mA

(all further figures are compensated for this load)

• 60litre Dometic Fridge with mini freezer switched to 12v : 9.5A
• Fridge cooing fans x2 : 300mA

• LED RGB strip-lights lighting controller: 100mA
• 7metre LED strip-lights set to 100% red: 1.4A
• 7metre LED strip-lights set to 100% red, 100% Green & 100% Blue (ie. full white-ish) :3.5A

• Overhead 12volt LED dichroics (6watt total): bow pair: 500mA
• Stern pair: 700mA (voltage drop on longer cable)
• Mid tungsten pair (40watt total): 3.3A

• Phone charger 5volt USB port: 300mA

• Water pump (continuous) 1.7A

• 600watt Modulated inverter idle (ridiculously over-spec.ed): 400mA
• Inverter: running 30watt 5.1 surround sound 1.0A
• Inverter: running aux. computer monitor: 2.9A
• Inverter: with mains laptop charger idle: 600mA
• Inverter: laptop battery charge only: 4.3A
• Inverter: laptop on (CPUs @ 10%) with laptop battery charging: 6.5A
• Inverter: laptop on (CPUs @ 10%) without laptop battery/fully charged laptop battery: 2.8A

• DC Laptop charger: 19volt 120watt (running @ 90watts max) idle:100mA (ish)
• DC Laptop charger: laptop battery charge only: 3.4A
• DC Laptop charger: laptop on (CPUs @ 10%) with laptop battery charging: 5.7A
• DC Laptop charger: laptop on (CPUs @ 10%) without laptop battery/fully charged laptop battery: 2.0A

Generation:

150watt solar module output absorption charge rate @17hr00, flat mount, covered in light road dirt/dust, sunny [EDIT: indirect ambient light (sun set behind house roof, but module yet unshaded)] 15°C: 4.8A

Some generation figures in;
(compensated for ghost loads)

Average overnight consumption:
12Ah - 20Ah

Daily generation figures are capped at what I use the previous night as the solar controller stops charging as the battery fills up.
I will have to run deeper discharges to get max. possible daily generation figures.


Days #1 & #2
12°-16° C
Sunny morning (direct sunlight) 09hr00: 3.3A
Cloudy morning (diffused sunlight) 09hr30: 2.6A

Midday sun through cloud (possible edging) 11:30: 5.0A

Overcast midday (Heavy rain-cloud): 1.8A
Bright Midday: 3.0A

Raining afternoon: 1.2A
Sunny evening 17hr10: 5.4A

Daily generation 20Ah.
Float Charge by 14:00 - 17:00

Day #3
max 20°C

Bulk Charge:

07hr00: 1.4A
Overcast cloudy 10hr00: 2.6A
Sun through light cloud (edging) 10hr30: 4.8A > 7.8A
Sun through cloud 11hr40: 4.0A
Sun through light cloud, (edging) 12hr05: 5.8A > 9.8A
Direct sunlight 12hr30: 7.5A
Direct sunlight 13hr40: 5.6A

Absorption:
Sun through light cloud 15hr30: 1.3A
Direct sun 19hr40: 200mA

No Float.
Aprox. Generation 22.5Ah

Day #4
17°C

Bulk Charge:

Sun through light cloud 11hr30: 5.5A
Sun through light cloud 11hr50: 4.1A

Absorption:
Sun through light cloud 13hr05: 1.3A
14hr25: 400mA

Fully charged - Float Charge @ 17hr00

Aprox. Generation 18Ah


I'm not using the batteries during the daytime at the moment, I'm preoccupied subjugating the tin worm.

The module I'm using is a monocrystalline as these outperform the polycrystallines in diffused light conditions, ie. cloudy weather. So are best suited to our climate.

The first two days of figures I used were when the module was covered in a light layer of road filth.
So just to satisfy my curiosity I climbed up the roof with some Jif Universal spray and gave the module a good seeing to. After it was clean and dry I re-checked the meter in what i reckoned was the same light conditions and I had gained 100mA > 200mA.

Normally I tend to charge all battery operated devices during the evening, after letting the leisure battery recharge to midday from the previous night. This means excess bulk charge is being directed and there's enough sunlight afterwards to complete an absorption charge cycle after my dry cell devices are charged. Also it means that the leisure battery is not being unnecessarily run down overnight for want of better organisation.

One of the experiments I was hoping to shed light on was Morningstar's bold claim that their PWM chargers can recover stratified batteries. I speculated in April that my last remaining battery was at 65% DOD after 2.5 years almost daily use up to 50% depletion (mostly) and weekly charging.
In early July after one month of PWM charge constant float I've calculated the same battery at 77% DOD
Now some of this I can attribute to the benefit of a temperature compensating charger but I'm still quite impressed
As of last night this seems to be on the rise again possibly from cycling the battery. Meter was showing 12.7v for 3 hours longer than the previous night with the same discharge.

As for my hopes of running the fridge on 12volt during the daytime I realised this was highly ambitious and impractical as the module will only ever produce 7.5A max in optimal conditions with a flat mount and the fridge consumes 9.5A. Compounding this, optimal conditions for a solar module are the least efficient for a fridge in which case I would have to use gas anyway.

I tried a deeper discharge last night but hit 12.3v on a 125Ah battery after 22.5Ah so I curtailed the endeavor. This likely means my DOD figures are way off...which is exactly what I should expect from using battery voltage to indicate % charge. I'll recalculate where I'm at using specific gravity instead one of these days...as I should have in the beginning :rolleyes:
The TriMetric does display % charge calculated by cumulative/subtractive amps in/out but as I set this to an optimistic 125Ah I don't deem this accurate either.

More Numbers


18°C - 25°C

Bulk Charge

Moderate Cloud 09hr00: 2.0A
Cloud Edging 09hr45: 3.3A > 5.3A
Direct Sun 10hr25: 5.8A
Direct Sun 10hr45: 6.2A
Heavy Cloud 11hr00: 1.6A
Heavy Cloud 12hr05: 2.8A

Absorption Charge

Direct Sun 12hr40: 3.0A
Applied load (fridge) 12:40: 7.9A

Direct Sun 13hr30: 1.5A
with load: 7.6A

Direct Sun 15hr00: 500mA
with load 7.2A

Light cloud & edging 15hr45: 400mA
with load 4.8A > 6.8A

Direct Sun 16hr30: 300mA
with load: 6.5A

Direct Sun 18hr00: 200ma
with load 5.0A

Overcast 18:50: 200mA
with load 1.5A

Direct Sun 19:55: 200mA
with load 1.3A

No Float...mostly because I kept switching loads on/off all day

Approx. Generation: 22.5Ah
Potential Max Generation (very rough approximation from above figures): 60Ah

Silicone paste works safely to 200°C

This is as far as I've got today.
Please let me know if anyone spots any glaring mistakes.

Battery #1 Status
Temperature 29°C
24 hours resting after full charge and watering followed by 4 hour equalisation charge.
Battery Voltage 13.0v


__________Specific Gravity_____S.G. @ 20°C_____Voltage
____________(± 0.010)______________________________
Cell #1________1.200____________1.202___________1.9v
Cell #2________1.230____________1.232___________2.2v
Cell #3________1.225____________1.227___________2.1v
Cell #4________1.200____________1.202___________2.2v
Cell #5________1.225____________1.227___________2.1v
Cell #6________1.210____________1.212___________2.4v

S/G at new 1.250 per cell (Elecsol just have to be different..)

Average Cell S.G. = 1.217
20% DOD

20% of 125Ah = 25Ah
Present Capacity = 100Ah

Battery #2 isn't looking so hopeful.
It's on it's second equalisation cycle now...5½ cells seems about right for what it is, from preliminary results. I ran out of photons to get it above 14.6v yesterday.
Probably not worth reinstating...hmmm...

paddyp wrote: »

I'd scrap it and put the money towards new one.

I totally agree, I'm just experimenting really.
Everybody kills their first batteries not worth getting emotional about it.

Re-instated and re-wired the bank today. Couldn't be bothered with the hydrometer but I will post comparative results in due time.
I've noticed definite improvement on the stronger battery (without giving Morningstar too much credit here I'm suspicious I ought to be reprimanding Halfords cheapo charger for an undercharging algorithm). I'd almost go so far as to say one battery on my current set-up is outperforming two at the end of my old (mains and alternator).
Bank is holding 12.9v after days charge.

I knocked 100mA off my ghost load by cleaning my battery terminals and terminal clamps. I thought the clamps were supposed to be grey because they were fully oxidised when I bought them. I was going to upgrade them for copper but seeing as it was a Sunday and I wasn't in the mood for a rip-off merchant, I just got the grinder out. Low and behold they're a shiney coppery/brass now and 0Ω.

I was getting 1Ω resistance across an inch of the clamp, this doesn't sound like a lot until you compare that to 0.3Ω from 20 metres of stranded 1.5mm copper cable. DC electrons travel across the surface area of the conductor so an oxidised surface is really killing it.

I know these numbers are very small but 100mA * 24 * 30 = dead battery at the end of a month without charge, before temperature induced loss and self-discharge.

I emailed Elecsol over a week ago requesting their recommended charge set-point voltages for their carbon fiber lead acids. They have not published this information anywhere I can find it. Nor can I find their charger algorithm set-points. No reply yet. :mad:

Morningstar on the other hand, you can email them a wiring diagram and they will get back to you in a few days with the pros and cons and practicalities.

Here's an idea I got indirectly from Morningstar that I have yet to implement, mostly because I thought I was going to soon be buying new batteries and I'm running low on 7mm cable. The Tristar controller has a 45A continuous rating (70A max for edging & spikes IIRC) I'm using ±9.5A from solar generation. If I was to link my alternator battery feed through the solar controller it will regulate the 14.7v (flooded battery set point) from the alternator to 14.4v (sealed battery set-point) for the leisure bank.

I wouldn't recommend this protocol on an automated relay charge set-up as it will always undercharge the starter. On a selective system this will work quite well.

[Edit:] Actually, second thought this could work on an automated system as second regulation is happening after the engine battery. This is not something I can speculate without measuring though. I wonder if there'd be back EMF issues? Might email Morningstar another diagram.

Any suggestions? Not looking in any particular direction Paddyp...{where's that innocent whistling smiley Peasant?...}

Wasn't relying on the clouds to charge the bank yesterday so I turned off the module and solar controller and stuck my mains charger on it. If I left the solar running it's piffle of amps at 14v it would fool the charger into thinking the batteries were charged, 'nuff said.

So low and behold my mains charger signs off on a charged bank and has a nice cup of tea. I check the Trimetric and see that it's 92% charged (92% of a 180Ah bank not 250Ah as the combined battery labels might indicate).

wrote:

Lead Acid Batteries and Undercharging.

Undercharging of the battery occurs when 107% to 115% of the removed ampere hours are not provided during the recharge. When not fully recharged, the residual lead sulphate remains on the positive and negative plates and eventually 'hardens'. With successive cycles of undercharging the layer of residual lead sulphate becomes thicker, the electrolyte specific gravity decreases, and the battery cycles down in capacity. In the 'hardened' condition, it may not be possible to convert the residual lead sulphate back into the original lead dioxide, sponge lead and sulphur acid active materials, even with higher voltage charging efforts. In this case the battery will suffer a permanent loss in capacity.

The Tristar frequently carries on charging a few hours beyond 100% at an increasingly reduced amperage.

The very fact that Halfords state that a 9Amp charger is suitable for a 200Ah battery (4.5% charge rate) should say it all really. Battery manufacturers recommend a 10%- 13% charge rate of the C20 Ah rating. Reasons like this are why I have such low esteem in battery charger manufacturer claims.
I will not be buying another mains charger unless it's temperature compensating, and programmable or the algorithm matches the battery manufacturer's guidelines exactly. No fixed algorithm charger (non-programmable) is "suitable for all battery types".

Have I convinced anyone to get a TriMetric yet? Will this help?

I started powering my fridge on photons, alternator and combined on the 12volt element, having never used it on 12v before.
The problem here is the engine battery isn't connected to the shunt (and that's unlikely to change soon) so the TriMetric tends to get befuddled from alternator charge as the electrons are skipping the turnstile and overestimates the used capacity. hmmm...maybe time to install an ammeter in the dash. In any case it self-resets when it detects a fully charged battery and in the short-term I just watch the volts instead.

Solar can power the fridge at about a 3A shortfall which works ok every second day to save the gas. How and ever my fridge has a 100watt element;
100wats /12.4 volts = 8A.

The TriMetric tells me it's running @ 9.2A @ > 13v.

Voltage drop methinks...there's 1.2A being used to heat cable back there. That'd be the 1.5mm cores I thought'll be grand a few years back then. So that's getting pulled out for 4mm then (2.5mm would be adequate on a run under 9metres but I'm a pedant when it comes to cable).

I read a suggestion which I think might be interesting to investigate. Which is that a 3-way absorption fridge works much better set to mains and powered off an inverter (modulated / RMS / pure sine wave...doesn't matter; it's a resistive element) than the 12volt setting as at 230V voltage drop is not an issue and the mains element has a thermostat. ...the thick plottens... once I have it asunder I'll see what the power consumption comparison is (+20% give or take I'm guessing...(inverter conversion loss and a 5watt stronger element) ...although not an issue with a large enough alternator).

Fridge Vent Fans;
Wonderful things! Fridge is way more efficient above 20°C with them running, for a humble 300mA (two fans).
Unfortunately my fans were 12volt. My nominal system voltage is now 13v, and usually 14v+ when I need them.
They have expired after putting up a good fight. I reckon I burnt the motor coils to bejaysus as at the end of their days I could hear them over-revving. It may just be a fried switch too but I've decided if I'm going in there then I might as well overhaul the lot (back of my fridge has access issues, if i knew when I was installing it what I know now then...ah..well...:rolleyes:).

I'm not a whole lot of upset about this because they were due and upgrade for brushless versions (quiet).

I had a notion of dual wiring them as generators (mmm permanent magnet motors:D) off the driving wind-speed with vent scoops but have abandoned the notion as a feeble return on investment and libelous over-engineering.

That just leaves me with the problem of running 12volt fans with a high voltage tolerance inexpensively on a 14volt variable system. Which I think I can solve simply with a strategic resistor.

Or running 24volt fans with a low tolerance at a significant reduction of torque.

Damn hard find decent fans with wide tolerences 'tis.

Ah cheap chineese kit
6 for the price of
¼ Quality made ones
tehehe...guess which ones I bought!